Wind turbine design is a challenging multidisciplinary optimization problem, where the aerodynamic shapes, structural member sizing, and material composition must all be determined and optimized. Some previous blade design methods incorporate static loading with an added safety factor to account for dynamic effects. Others incorporate dynamic loading, but in general limit, the evaluation to a few design cases. By not fully incorporating the dynamic loading of the wind turbine, the final turbine blade design is either too conservative by overemphasizing the dynamic effects or infeasible by failing to adequately account for these effects. We propose an iterative method that estimates fatigue effects during the optimization process while quickly converging to the true solution. We also demonstrate an alternate approach where a surrogate model is trained to efficiently estimate the dynamic loading of the wind turbine in the design process. This surrogate model, once trained, was then incorporated in the optimization loop of the wind turbine blade. In contrast to the iterative method, there is significant upfront computational cost to construct the surrogate model. However, this surrogate model has been generalized to be used for different rated turbines and can predict the fatigue damage of a wind turbine with less than 5% error for baseline wind turbines of the same family. These methods can be used instead of the more computationally expensive method of calculating the dynamic loading of the turbine within the optimization routine.